“Riemannian Geometry” is maybe less sexy than “Deep Learning” 🤔
But yesterday, we had up to 90 researchers and students at our workshop for the Brain Computer Interface Society #vBCI2021. 🚀
You couldn’t make it?! 😥
No worry, all the resources are available on Github 😁. You can find slides and code examples to learn how to:
– Automatically detect the artifacts in your data using the Riemannian Potato 🥔 (much better than French fries but still cooked with 100% French researchers 🇫🇷).
– Upgrade your sklearn classification pipeline with a grain of Riemannian Geometry for better results. 🧂
– Benchmarking ALL the classification methods you can imagine on your BCI datasets using the Mother Of All BCI Benchmark 🙌 (a.k.a MOABB, and yes, it is a real name)
– Use all of the above methods on your real-time BCI system using Timeflux. 🧠
All resources are open-sources!
One more thing:
Marco Congedo ask me to invite any researcher, professor, post-doc or student to come visit GIPSA-lab (UGA, Grenoble, France) if you want to start a collaboration on Riemannian Geometry applied to BCI. Contact me if you are interested.
Thanks for the speakers: Sylvain Chevallier, Florian Yger, Pierre Clisson, and Marco Congedo.
Special thanks to Quentin Barthélemy & Alexandre Barachant for sharing these amazing open-science tutorials on Pyriemann and Alexandre Gramfort for maintaining the repo.
Here is the resources you need! Don’t hesitate to fork and please let us know how it went. You can create an issue if you need help.
https://github.com/lkorczowski/BCI-2021-Riemannian-Geometry-workshop